I've had my thoughts for a while on the usage of GitHub issues for open source projects. This is going to sound weird but I'm actually for a bit more gatekeeping in the form of a different system such as bugzilla. When you weigh "what if we miss a report" against the pure volume of issues that come in against how many of these issues can be realistically handled is it really worth it?

Expectations of developer AI usage: "AI, go make me this code"
Reality: "AI, tell me which of these 300 GitHub issue comments is a workaround for my problem"

UI is very important in terms of open source software adoption. Sure it's fun as a programmer to see how much you can get away with using ncurses on terminal only, but at the end of the day normal users would probably like to just click the shiny button with obvious intent presented.

I miss floppy disks. Obviously the storage is abysmal by today's standards, but putting one into a floppy drive felt so satisfying. Then you had those wonderful floppy drive reading sounds. Now you have USB thumb drives where you kind of lightly bump it into the port and data transfer is silent. It just don't feel hacker enough!

There's a certain segment of tech recruiters who think that all programmers magically know Java. It's kind of weird...

I like how they call it a kernel panic because let me tell you I'm certainly panicking more than the kernel when it happens

Some days you just feel like searching for "Open Source Warhammer 40k" to test fate itself.

I played with legos a lot as a kid does that count as "Assembly Expert" for a microcontroller programming position?

C as a language is pretty simple. It's just you add in the stdlib, then kernel calls, then you want to support 3 different operating systems, with multiple architectures, and oh you want to support musl as well as glibc? There's a lot of rabbit holes down there.